基于模糊滚动RRT算法移动机器人路径规划.pdfVIP

34 5 () ol. 34 No. 5 2010 10 JournalofNanjing University of Science and Technology (NaturalScience) Oct. 2010 RRT 康 亮, 赵春霞, 郭剑辉 ( , 210094) : 研究了环境未知情况下的移动机器人路径规划问题, 将快速扩展随机树 (RRT)算法 基于滚动窗口的路径规划相结合, 提出一种新的移动机器人路径规划算法, 克服了 RRT算法通 常只能在已知环境中进行移动机器人路径规划的限制规划时只考虑窗口环境地图, 提高了 RRT算法规划效率,保证了算法的实时性针对 RRT算法路径规划缺乏确定性的问题, 结合人 类经验及模糊控制理论, 以概率来随机取点,并引入启发式估价函数,使随机树易于朝目标点方 向生长同时运用回归分析生成新节点, 增强了算法搜索未知空间的能力, 避免了可能产生的 局部极小最后仿真实验验证了该算法的有效性 : 移动机器人;路径规划;模糊控制; 滚动规划; 快速扩展随机树 : TP 24 : 1005- 9830( 2010) 05- 0642- 07 Path Planning Based on FuzzyRollingRapidly-exploring Random Tree forMobileRobot KANG L iang, ZHAO Chun-xia, GUO Jian-hui ( Schoolof Computer Science and Technology, NUST, Nanjing 210094, China) Abstract: Themobile robot path planning in an unknown environment is studied. The rapidly-ex- ploring random tree (RRT) algorithm is combined w ith the rolling path planning in the planning, and anovelpath planning isproposed. It is performed on-line in a rolling style, so that theRRT a-l gorithm can be used in path planning not only when the environment is known but also unknown. Only the local environm entmap is calculated in planning so as to mi prove the planning efficiency, and the real tmi e is guaranteed. Ow ing to lack of certainty in path planning of RRT, the random configuration using both fuzzy control theory and human experience is forced to coincidew ith the de- sired areaw ith some finiteprobability. The algorithm has introduced theheuristic evaluation function so that the exploring random treew ill grow in the direction of target point. The regression analysis, which avoids localm inmi a, enhances the capability of searching an unknown space. Them ethod ef- fectiveness is v